Need supplier, error-proof inspection of ID threads present attribute

[apestate]

Hello everyone

We have a situation that may require a goofy step along the path to a solution.

We are producing a nut with M8 threads inside. The nut is 17.7 mm long and made from 1/2" hex. The threads are cut threads, as opposed to rolled or formed threads.

We need to check these parts for presence of threads with some design for error-proofing. I have many ideas for a good machine to do this job, but am not getting the OK to build it.

In the interim, is there such a thing as a supplier of error proof inspection? We are looking for a company to take these parts and run them through some process that will weed out parts that do not have threads.

Any ideas?

 

[Wes Bucey]

Hello everyone

We have a situation that may require a goofy step along the path to a solution.

We are producing a nut with M8 threads inside. The nut is 17.7 mm long and made from 1/2" hex. The threads are cut threads, as opposed to rolled or formed threads.

We need to check these parts for presence of threads with some design for error-proofing. I have many ideas for a good machine to do this job, but am not getting the OK to build it.

In the interim, is there such a thing as a supplier of error proof inspection? We are looking for a company to take these parts and run them through some process that will weed out parts that do not have threads.

Any ideas?

Eric,
Something seems amiss here. It appears you intend "quality by detection" versus "Quality by prevention."

What happens to generate a threadless piece? Systemic failure or random?

Is the thread cut on same machine which cuts off hex or is it a secondary?

Are the parts threaded in a secondary operation?

I used a Nakamura turning center (two identical machines) to turn out millions of similar hex parts with internal tapered thread (much more complicated external machining - external thread, etc. - than a simple nut, complete in one pass from bar stock) and used only in-process inspection to prevent any nonconforming parts from reaching customer.

 

[D.Scott]

Hello everyone

We have a situation that may require a goofy step along the path to a solution.

We are producing a nut with M8 threads inside. The nut is 17.7 mm long and made from 1/2" hex. The threads are cut threads, as opposed to rolled or formed threads.

We need to check these parts for presence of threads with some design for error-proofing. I have many ideas for a good machine to do this job, but am not getting the OK to build it.

In the interim, is there such a thing as a supplier of error proof inspection? We are looking for a company to take these parts and run them through some process that will weed out parts that do not have threads.

Any ideas?

Eric,

Please check our site www.forestcitytech.com

We have been doing vision sorting for a number of years. I'm sure we could help.

Dave

 

[apestate]

Wes

we are running New Britain multi-spindles model 51 and 62. you can get a really swank NB multi for $60,000.

the parts we make get about $0.40 after plating, and we make about 1/2 million a year of this nut.

I recognize the wisdom in reducing dependance on inspection, if we could, we would. The machine would have to be modified to some sort of servo driven rigid tapping to accomplish this.

It's a screw machine. They're good at making simple parts rapidly. They are not good at keeping oil inside the machine and this particular setup is not good at threading 100% of the time. It's kind of a rickety and unreliable process.

I try to fit these circumstances into quality disciplines and philosophies but end up salivating at the idea of weighing the parts individually at high speed. It's a cut thread, you see.

Thank you for your responses. It's a difficult situation, as you see, because the part is so entirely basic and inexpensive. The presence of threads, however, seems extremely valuable to the customer.

 

[Wes Bucey]

Wes

we are running New Britain multi-spindles model 51 and 62. you can get a really swank NB multi for $60,000.

the parts we make get about $0.40 after plating, and we make about 1/2 million a year of this nut.

I recognize the wisdom in reducing dependance on inspection, if we could, we would. The machine would have to be modified to some sort of servo driven rigid tapping to accomplish this.

It's a screw machine. They're good at making simple parts rapidly. They are not good at keeping oil inside the machine and this particular setup is not good at threading 100% of the time. It's kind of a rickety and unreliable process.

I try to fit these circumstances into quality disciplines and philosophies but end up salivating at the idea of weighing the parts individually at high speed. It's a cut thread, you see.

Thank you for your responses. It's a difficult situation, as you see, because the part is so entirely basic and inexpensive. The presence of threads, however, seems extremely valuable to the customer.

Bummer! Doesn't seem to be a lot of value added on this item for machining. What do you do for other orders for internal threads?

At 40 cents gross price, you need to know what happens at the customer. If he is installing versus loading into a kit, he should agree to merely discard threadless nuts and give you a weekly/monthly tally for financial settlement. It will result in 100% detection and be cheaper for you in the long run than paying for inspection, sorting, AND discarding. (Customer throws discards into a container for weighing instead of counting. He scraps at his end, no shipping of bad parts back to you.)

I note nothing seems to be mentioned about quality of threads that do get cut. How is that working out?

 

[Miner]

Here is an article on the merits of several 100% thread inspection alternatives. Just keep in mind that the company is trying to sell their system.

https://www.newvistacorp.com/tvu_1_more.htm

 

[apestate]

thanks everyone

Wes, that would probably be closer to a logical arrangement. We have a pretty good quality reputation, you see, because of a little bit of discipline on the floor. We work to remove bar ends and short parts that inevitably get created.

The presence of threads is the #1 mission critical aspect of this part. If the threads are poor quality, to some extent they can handle that. I don't understand the situation fully, but apparently a line that builds tractors goes down every time we make a threadless 40 cent nut. If I recall, they have to tear down the tractor somewhat to replace the part, and they order exactly how many parts they need.

They cannot detect the threads absent fault. The parts are automatically injection molded into a plastic assembly and the threads don't get used until the parts are at our customer's customer.

Our customer and our customer's customer is asking us for some sort of error proof thread presence design in our process to eliminate this fault. I'm doing research to build a line to weigh them one by one at .01 gram resolution, 20 parts per minute.

Thank you for your discussion Wes. Thank you, also, Miner and D. Scott. I'm just starting the research to solve this problem and any bit of help is greatly appreciated.

 

[True Position]

In my humble opinion, your weight based system will have lots of false alarm issues or acceptance of NC material. My only previous experience with automated tapping was running a production pierce nut line. They had an optical system that each part would run under, and if a non-tapped hole was found the line would stop. If you could easily run each part under an optical/laser eye this type of setup may be better for you. The system was triple redundant if I recall, using three eyes set to different locations along the thread.

Something optical or eddy current(probably too slow for your needs) would seem more approprate for your requirements. Have you weighed a few hundred of your fasteners to be sure there's a discernable difference? Chips, size of OD, coolant/oil, and many other things I'm sure I'm not thinking about can fool a weight based system.

Here's an example of something similar to what I was talking about:
https://www.mectron.net/systems_n250.cfm

 

[Trae1170]

Hello everyone, I am pretty new to the forum (a few days) but some options to help detect your probllem at the source & limit the Qty. of scrap:

Some type of broken tool detection in your machine (probe or laser with air blow-off - I know a company named BLUM makes this)

A two-pan system at the machine - At our company we have about 15 multi-spindles also and because of the amount of pieces that can be generated in a short period of time we have all operators utilize the 2 pan system. As parts come off of the machine they go into the first pan, when this pan is full they must fully inspect one piece from each spindle (all 6), if these pieces are good then they empty the parts into the final pan in which they will be sent to the next operation in. If they find an issue (in your case, missing threads), they stop the machine, replace the tap & drill if needed & then run out all the scrap parts in the remaining spindles. Now the operator has the scrap parts contained in a smaller amount that will need to be sorted, you should never have any parts get past this operation with missing threads, as long as you have a robust procedure in place for "Discrepant Material", telling your operators what to do if scrap material is generated.

I hope this helps,

 

[Tom Thread]

Are you still in need of a solution for your quality issue?

Thanks,

Tom Thread